Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Like the title says I'm attempting to write a existence proof showing "that there exists no non-zero real numbers a and b such that $\sqrt{a^2 + b^2} = \sqrt[3]{a^3 + b^3}$".

I'm having trouble finding a starting point. I've tried manipulating the equation but haven't managed to isolate a or b. But I don't know if this helps me or where to go from there.

If anybody could give a hint I would very much appreciate it.

P.S. I would ask that you refrain from posting a full solution as I'm trying to learn this material.

share|improve this question
1  
There is a LaTeX symbol, \neq, for $\neq$. –  Dustan Levenstein Jan 10 '12 at 4:14
    
@Dustan Levenstein thanks I couldn't remember it –  AvatarOfChronos Jan 10 '12 at 4:21
    
There is also \not = for $\not =$, which I find easier to remember –  Henry Jan 10 '12 at 8:22

1 Answer 1

up vote 14 down vote accepted

Hint: Take the sixth power of both sides.

Added: (for completeness, after the OP had used the hint to solve the problem) Since $\sqrt{a^2+b^2}$ is by definition non-negative, we have $$\sqrt{a^2+b^2}=\sqrt[3]{a^3+b^3} \qquad\text{iff}\qquad \left(\sqrt{a^2+b^2}\right)^6=\left(\sqrt[3]{a^3+b^3}\right)^6.$$ Equivalently, we want to find the solutions of $$(a^2+b^2)^3=(a^3+b^3)^2.$$ in non-zero reals. Expand. We get $$a^6 +3a^4b^2+3a^2b^4+b^6=a^6+2a^3b^3+b^6,$$ which is equivalent to $$3a^4b^2+3a^2b^4=2a^3b^3.$$ Since we are looking for solutions where neither $a$ nor $b$ is equal to $0$, we are looking for non-zero real solutions of $$3a^2+3b^2=2ab.$$ This equation has no non-zero real solutions. For by completing the square we get $$a^2+b^2-\frac{2ab}{3}=\left(a-\frac{b}{3}\right)^2+\frac{8b^2}{9}.$$ In order for the right hand side to be $0$, both terms must be $0$. In particular, $b$ must be $0$, and therefore so must $a$.

Comment: You may find the following related idea interesting. Let $p>1$, and let $t$ be positive. We will prove that $$1+t> (1+t^p)^{1/p}. \qquad\qquad(\ast)$$ Let $f(t)=1+t -(1+t^p)^{1/p}$. Note that $f(0)=0$. So it is enough to show that for positive $t$, $f(t)$ is an increasing function. To do this, we use the derivative: $$f'(t)=1 -\frac{t^{p-1}}{(1+t^p)^{(p-1)/p}}.$$ For positive $t$, $f'(t)$ is positive, since $(1+t^p)^{(p-1)/p}>(t^p)^{(p-1)/p}=t^{p-1}$. This completes the proof of $(\ast)$.

To apply $(\ast)$ to our problem, observe first that a solution of our equation with positive $a$ yields a solution of $(1+(b/a)^2)^{1/2}=(1+(b/a)^3)^{1/3}$. It is easy to check that $b/a$ cannot be negative. Now let $t=(b/a)^2$. We obtain the equation $1+t=(1+t^{3/2})^{2/3}$, which, by $(\ast)$, cannot hold for positive $t$.

We reduced the problem to one variable in order to use familiar tools. But there are important generalizations to several variables.

share|improve this answer
    
had not thought of that one. Trying it now. –  AvatarOfChronos Jan 10 '12 at 4:15
    
Many thanks for your hint and your timely reply, I have my solution and I think it's correct. –  AvatarOfChronos Jan 10 '12 at 4:28
    
This would have been better suited as a comment. –  user18063 Jan 10 '12 at 4:57
3  
$3a^2+3b^2-2ab=2a^2+2b^2+(a-b)^2$. –  Did Jan 10 '12 at 6:07
    
Both sides are homogeneous...so WLOG we can consider $a=1$. –  Tapu Jan 10 '12 at 6:11

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.